Update app.py

app.py

@@ -16,38 +16,38 @@ depth_model = DPTForDepthEstimation.from_pretrained("Intel/dpt-large")
 def apply_gaussian_blur(image):
     # Resize and preprocess the image
     image = image.resize((512, 512)).convert("RGB")
-    # inputs = image_processor(image, return_tensors="pt")
+    inputs = image_processor(image, return_tensors="pt")
 
-    # # Perform semantic segmentation using the model
-    # with torch.no_grad():
-    #     outputs = model(**inputs)
-    #     logits = outputs.logits
+    # Perform semantic segmentation using the model
+    with torch.no_grad():
+        outputs = model(**inputs)
+        logits = outputs.logits
 
-    # # Get the predicted class for each pixel
-    # segmentation = torch.argmax(logits, dim=1)[0]  # Shape: [height, width]
+    # Get the predicted class for each pixel
+    segmentation = torch.argmax(logits, dim=1)[0]  # Shape: [height, width]
 
-    # # Create a binary mask for the 'person' class
-    # person_index = 12  # Assuming 12 is the 'person' class index
-    # binary_mask = (segmentation == person_index).numpy().astype(np.uint8) * 255  # Convert to 0 and 255
+    # Create a binary mask for the 'person' class
+    person_index = 12  # Assuming 12 is the 'person' class index
+    binary_mask = (segmentation == person_index).numpy().astype(np.uint8) * 255  # Convert to 0 and 255
 
-    # # Convert the original image to a numpy array
-    # image_np = np.array(image)
+    # Convert the original image to a numpy array
+    image_np = np.array(image)
 
-    # # Apply Gaussian blur to the entire image
-    # blurred_image = cv2.GaussianBlur(image_np, (0, 0), sigmaX=15, sigmaY=15)
+    # Apply Gaussian blur to the entire image
+    blurred_image = cv2.GaussianBlur(image_np, (0, 0), sigmaX=15, sigmaY=15)
 
-    # # Normalize the mask to range between 0 and 1
-    # normalized_mask = binary_mask / 255.0
-    # normalized_mask = np.expand_dims(normalized_mask, axis=-1)  # Add channel dimension
+    # Normalize the mask to range between 0 and 1
+    normalized_mask = binary_mask / 255.0
+    normalized_mask = np.expand_dims(normalized_mask, axis=-1)  # Add channel dimension
 
-    # # Create the composite image with the blurred background
-    # final_image = (image_np * normalized_mask + blurred_image * (1 - normalized_mask)).astype(np.uint8)
+    # Create the composite image with the blurred background
+    final_image = (image_np * normalized_mask + blurred_image * (1 - normalized_mask)).astype(np.uint8)
 
-    # # Convert back to PIL Image
-    # final_image_pil = Image.fromarray(final_image)
+    # Convert back to PIL Image
+    final_image_pil = Image.fromarray(final_image)
 
-    # return final_image_pil
-    return image
+    return final_image_pil
+    # return image
 
 def apply_lens_blur(image):
     # Resize and preprocess the image